Upload
alvis
View
58
Download
1
Tags:
Embed Size (px)
DESCRIPTION
Indo-Australia Workshop on Optimization in Human Language Technology 16 th Dec 2012, IIT Patna. Language Change as a Constrained Multi-Objective Optimization. Monojit Choudhury Microsoft Research Lab, India [email protected]. A tale of the lazy tongue. Language Change. - PowerPoint PPT Presentation
Citation preview
Language Change as a
Constrained Multi-Objective
Optimization
Monojit ChoudhuryMicrosoft Research Lab, India
A tale o
f the la
zy ton
gue
Indo-Australia Workshop on Optimization in Human Language Technology16th Dec 2012, IIT Patna
Language Change
Language Change
• Change in the syntactic/semantic/phonological features of a language
• Perpetual, universal, directional (?)
• Phonological Change: – Affects the sounds– Structured, independent of syntax/semantics– Example: Loss of consonant clusters in Hindi
agni aag, dugdha dUdh, raatri raat
Effects of the “Lazy Tongue”
Assimilation• in+apt = inapt• in+decent = indecent• in+polite = impolite• in+mature = immature• in+legal = illegal• in+regular = irregular
Deletion• cannot can’t• do not don’t• will not won’t• are not ain’t• information info
Explanations for Change
Exogenous causes– Language contact– Socio-political
factors– Communication
medium
Endogenous causes– Functional– Phonetic error-based– Frequency drifts– Evolutionary
Functional Explanation of Language Change
• There are three evolutionary forces on any linguistic system:– Minimization of effort (energy)– Maximization of perceptual distinctiveness
(Minimization of ambiguity)– Maximization of learnability
Language is a perpetually evolving system shaped by these three conflicting
forces
Outline of the Talk
• Morpho-phonological change of Bangla Verb systems and emergence of dialect diversity– Approach: Multi-Objective Constrained Optimization– Technique: Multi-Objective Genetic Algorithm (MOGA)
• Understanding Computer Mediated Communication– Normalization of Texting language – Romanization of Indian Language text
Geography of Bangla
• Standard Colloquial Bengali (SCB)
• Agartala Colloquial Bengali (ACB)
• Sylhetti
History of Bangla
1200 AD 1800 AD
BanglaVerb Morphology
করে�ছি�লামkar-echh-il-aam
Verb root (do)
Aspect (perfect)
Tense (past)
Person (first)
I had done
Cognates in the Dialects
Features Classical SCB ACB
Non-finite kariyA kore kairAPs,2, per. kariyAChila koreChilo korsiloPs,1, cont. kariteChilAm korChilAm kartAslAm
root: kar (to do)
Atomic Phonological Operators
kariteChila
kariChila
kairChila korChila
karitChila
korChilo
Del(e/t_Ch)
Del(t/_Ch)Met(ri/_Ch)
Asm(ao/_i) Mut(a o/_$)
Deletion, MetathesisAssimilation, Mutation
Hypothesis
A sequence of Atomic Phonological Operators, is preferred if the verb forms obtained by application of this sequence on the classical forms have some functional benefit over the classical forms.
Thus, all the modern dialects of Bangla have some functional advantage over the classical dialect.
A Formal Model of Functional Explanation
f1: Effort of articulation
f2: [Acoustic distinctiveness]-1
Unstable languages
Impossible languages
Metastable languages
Genetic Algorithm
Gene (A string of symbols) How the solution actually looks like
GA: search for good solutions mimicking nature [recombination and mutation of genes]
Phenotype
kori
korChi
:
korte
kori
kartAsi
:
kartA
Lexicon consisting of 28 forms for the verb kar
Genotype
A sequence of atomic phonological operators
Del t Met ri NOP Del e Asm a Del i NOP
Dsm e NOP NOP Met ri Asm a Del e NOP
Genotype Phenotype
karikariteChi
karite
Del t Met ri NOP Del e Asm a Del i NOP
karikarieChi
karie
kairkaireChi
kaire
korkorCh
kor
Crossover
Mutation
Multi-Objective GA
Multi-Objective GA: Apply constraints
Multi-Objective GA: Apply constraints
Multi-Objective GA: Finding out good solutions
Multi-Objective GA: But also keep some not-so-good solutions
Multi-Objective GA: But also keep some not-so-good solutions
Multi-Objective GA: After several iterations
Objective functions
• Articulatory effort– fe(Λ): weighted sum of number of syllables,
letters and vowel height differences averaged over all words in the lexicon
• Acoustic Distinctiveness– fd(Λ): Inverse of mean edit distance between
words• Learnability
– fr(Λ): correlation between feature match and edit distance
Experiments
• NSGA – II : a package for fast MOGA• Gene length: 15 APOs• A repertoire of 128 APOs• Population: 1000, Generation: 500• 6 Models with different combinations of
constraints and objectives
Pareto-optimal front
CB
SylhettiACB
SCB
Observations
• vertical and horizontal limb• real dialects on the horizontal limb• Sound changes push the dialects from right
to left (reduce effort)• but never up the limb• why?
Role of Constraints
For more information
Choudhury et al., Evolution optimization and language change: the case of Bengali verb inflections, in Proceedings of ACL SIGMORPHON9, Association for Computational Linguistics, 2007http://research.microsoft.com/people/monojitc/
MOGA and NSGA IIKanpur Genetic Algorithms Laboratoryhttp://www.iitk.ac.in/kangal/index.shtml
Food for Thought
• Evaluation:– Myriads of possible dialects, but only a few
observed in nature• Fixed set of pre-defined APOs – how to
generalize for any change?
• MOGA is an optimization tool, which in no way simulates language change– How do languages optimize themselves?
Outline of the Talk
• Morpho-phonological change of Bangla Verb systems and emergence of dialect diversity– Approach: Multi-Objective Constrained Optimization– Technique: Multi-Objective Genetic Algorithm (MOGA)
• Understanding Computer Mediated Communication– Normalization of Texting language – Romanization of Indian Language text
Computer Mediated Communication
Form
Texting Language• A new genre of English & also other
languages used in chats, sms, emails, blogs, tweets, FB posts, comments etc.
dis is n eg 4 txtin lang
This is an example for Texting language
Texting Language• A new genre of English & also other languages
used in chats, sms, emails, blogs, etc.• Ungrammatical, unconventional spellings
dis is n eg 4 txtin lang
This is an example for Texting language
24 39
The shorter the fasterConstraint: understandability
Analysis of Social Media
• A hot topic in NLP– Normalization– Language identification– Sentiment/Polarity detection– Summarization/trend prediction
Choudhury et al. (2007) Investigation and Modeling of the Structure of Texting Language. In IJCAI Workshop on Analytics of Noisy Data 2007
Tomorrow never dies!!!
• 2moro (9)• tomoz (25) • tomoro (12) • tomrw (5)• tom (2)• tomra (2)• tomorrow (24)• tomora (4)
• tomm (1)• tomo (3)• tomorow (3)• 2mro (2)• morrow (1)• tomor (2)• tmorro (1)• moro (1)
Patterns or Compression Operators
• Phonetic substitution (phoneme)– psycho syco, then den
• Phonetic substitution (syllable)– today 2day , see c
• Deletion of vowels– message mssg, about abt
• Deletion of repeated characters– tomorrow tomorow
Patterns or Compression Operators
• Truncation (deletion of tails)– introduction intro, evaluation eval
• Common Abbreviations– Bangalore blr, text back tb
• Informal pronunciation– going to gonna, better betta
HMMs for SMS Normalization
G1
‘T’
S6
G2
‘O’G3
‘D’G4
‘A’G5
‘Y’
S0P2
/AH/P4
/AY/
S1
“2”
ε T @ ε O @ ε D @ ε A @ ε Y @
Bigram Examples
• TL: would b gd 2 c u some time soon• Op: would be good to see you some time soon
• TL: just wanted 2 say a big thanx 4 my bday card• Op: just wanted to say a big thanks for my today
card
• TL: me wel i fink bein at home makes me feel a lot more stressed den bein away from it
• Op: me well i think being at home makes me feel a lot more stressed deny being away from it
Code mixing
Transliteration
Spelling Change
Indian English
Use of Indian Languages on Online Social Media
Concluding Remarks
• Languages are perpetually evolving and optimizing systems– Computational modeling of language change is
still in its infancy– Lots of scope for research
Thank [email protected]
Questions??
Why Computational Models?
FOR AGAINST
FormalizationVirtual experimentation
Exploration
IntractableSimplified assumptions
Toy languages
Can we modelreal world language change?
Objectives and Constraints - 1
• Articulatory effortfe(w) = α1 fe1(w) + α2 fe2(w) + α3 fe3(w)
fe1(w) = |w|fe2(w) = hr(σi)
fe3(w) = |ht(Vi) - ht(Vi+1)|
Objectives and Constraints - 2
• Acoustic distinctivenessfd(Λ) = (1/N) ed(wi,wj)-1
Cd(Λ) = -1 if ed(wi,wj) = 0 for > 2 pairs
• Phonotactic constraints Cp(Λ) = -1 if any of the words violate
the phonotactic constraints of the language
Objectives and Constraints - 3
• Learnability as Regularity– fr: The correlation coefficient between the edit
distance and number of matching morphological attributes for every word pair
– Cr = -1 if fr > 0.8
Emergent dialects
Classical D1 D2 D3
kariteChilAm kartA karChi(korChi)
karteChi(kartAsi)
kariteChila kartAa karCha(korCha)
karteCha(kartAsa)
kariteChilen kartAen karChen(korChen)
karteChen(kartAsen)